COVID-19 Progression: Australian Data Only

library(tidyverse)
library(rlang)
library(lubridate)
library(scales)
library(ggrepel)
library(glue)
library(rvest)
library(pander)
library(plotly)
library(QuantTools)
panderOptions("big.mark", ",")
panderOptions("table.split.table", Inf)
panderOptions("table.style", "rmarkdown")
panderOptions("missing", "")
theme_set(theme_bw())

auStates <- c(
  ACT = "Australian Capital Territory",
  QLD = "Queensland",
  NSW = "New South Wales",
  VIC = "Victoria",
  SA = "South Australia",
  WA = "Western Australia",
  NT = "Northern Territory",
  TAS = "Tasmania"
)

Data Sources

getCovidTable <- function(
  state = c("ACT", "QLD", "NSW", "VIC", "SA", "WA", "NT", "TAS"),
  type = c("cases", "recoveries", "deaths", "active-cases", "tests"),
  .state_full = auStates
){
  
  state <- match.arg(state)
  
  type <- match.arg(type)
  type_name <- c(
    cases = "confirmed",
    recoveries = "recovered",
    deaths = "deaths",
    `active-cases` = "active",
    tests = "tests"
  )[[type]]
  col_name <- c(
    cases = "CASES",
    recoveries = "RECOV",
    deaths = "DEATHS",
    `active-cases` = "ACTIVE",
    tests = "TESTS"
  )[[type]]

    url <- glue("https://covidlive.com.au/report/daily-{type}/{str_to_lower(state)}")
  url %>%
    read_html() %>%
    html_nodes("body") %>%
    xml_find_all(
      glue("//table[contains(@class, 'DAILY-{str_to_upper(type)}')]")
    ) %>% 
    html_table() %>%
    .[[1]] %>%
    as_tibble() %>%
    dplyr::select(date = DATE, !!sym(type_name) := !!sym(col_name)) %>%
    separate(date, into = c("day", "month")) %>%
    mutate(
      year = case_when(
        month %in% c("Jan") & day < 25 ~ 2021L,
        TRUE ~ 2020L
      ),
      date = paste(year, month, day, sep = "-"),
      date = parse_date_time(date, orders = "%Y-%B-%d"),
      date = ymd(date),
      State = .state_full[str_to_upper(state)],
      Country = "Australia"
    ) %>%
    mutate_at(
      .vars = type_name, 
      .funs = function(x){as.integer(str_remove_all(x, ","))}
    ) %>%
    dplyr::select(State, Country, date, !!sym(type_name)) %>%
    arrange(date) 
}

confirmed <- names(auStates) %>%
  lapply(getCovidTable, type = "cases") %>%
  bind_rows()

recovered <- names(auStates) %>%
  lapply(getCovidTable, type = "recoveries") %>%
  bind_rows()

deaths <- names(auStates) %>%
  lapply(getCovidTable, type = "deaths") %>%
  bind_rows()

active <- names(auStates) %>%
  lapply(getCovidTable, type = "active-cases") %>%
  bind_rows()

tested <- names(auStates) %>%
  lapply(getCovidTable, type = "tests") %>%
  bind_rows() %>%
  dplyr::filter(!is.na(tests))

getLocal <- function(  
  state = c("ACT", "QLD", "NSW", "VIC", "SA", "WA", "NT", "TAS"),  
  .state_full = auStates
){
  
  state <- match.arg(state)
  
  url <- glue("https://covidlive.com.au/report/daily-source-overseas/{str_to_lower(state)}")
  url %>%
    read_html() %>%
    html_nodes("body") %>%
    xml_find_all("//table[contains(@class, 'DAILY-SOURCE-OVERSEAS')]") %>% 
    html_table() %>%
    .[[1]] %>%
    as_tibble() %>%
    dplyr::select(date = DATE, local = LOCAL) %>%
    separate(date, into = c("day", "month")) %>%
    mutate(
      year = case_when(
        month %in% c("Jan") & day < 25 ~ 2021L,
        TRUE ~ 2020L
      ),
      date = paste(year, month, day, sep = "-"),
      date = parse_date_time(date, orders = "%Y-%B-%d"),
      date = ymd(date),
      State = .state_full[str_to_upper(state)],
      Country = "Australia",
      local = str_remove_all(local, ","),
      local = as.integer(local)
    ) %>%
    dplyr::select(State, Country, date, local) %>%
    arrange(date) 
}
local <- names(auStates) %>%
  lapply(getLocal) %>%
  bind_rows() %>%
  dplyr::filter(!is.na(local))

dt <- max(confirmed$date)
if (hour(Sys.time()) < 9) dt <- dt -1
dt_char <- as.character(dt)

vic_aged <- "https://covidlive.com.au/report/daily-outbreaks-aged" %>%
  read_html() %>%
  html_nodes("body") %>%
  xml_find_all("//table[contains(@class, 'DAILY-OUTBREAKS-AGED')]") %>% 
    html_table() %>%
    .[[1]] %>%
    as_tibble() %>%
    dplyr::select(date = DATE, deaths = DEATHS, active = ACTIVE) %>%
    separate(date, into = c("day", "month")) %>%
    mutate(
      year = case_when(
        month %in% c("Jan") & day < 25 ~ 2021L,
        TRUE ~ 2020L
      ),
      date = paste(year, month, day, sep = "-"),
      date = parse_date_time(date, orders = "%Y-%b-%d"),
      date = ymd(date),
      State = "Victoria (Aged Care)",
      Country = "Australia"
    ) %>%
  arrange(date) %>%
  mutate(
    deaths = na_locf(deaths),
    active = na_locf(active),
    new = c(active[[1]], diff(active)),
    new = ifelse(new < 0, 0, new),
    confirmed = cumsum(new),
    recovered = confirmed - deaths - active,
    recovered = ifelse(recovered < 0, 0, recovered)
  ) %>%
  dplyr::filter(date <= dt) %>%
  dplyr::select(
    State, Country, date, deaths, confirmed, active, recovered
  ) 

Data for confirmed cases, active cases, recoveries and fatalities was exclusively sourced from COVID LIVE. Reliable data sources were extremely challenging prior to April and as such, minimal data is available from the early stages of the outbreak.

Similarly, data for the Victorian Aged Care outbreak was obtained from data based on press releases.

International Data

International data and figures can be viewed here

Latest Australian Data

ausPops <- tribble(
  ~State, ~Population,
  "New South Wales",    8117976,
  "Victoria", 6629870,
  "Queensland", 5115451,
  "South Australia", 1756494,
  "Western Australia", 2630557,
  "Tasmania", 535500,
  "Northern Territory", 245562,
  "Australian Capital Territory", 428060
)

Australian State populations were taken from the ABS Website and were accurate in Sept 2019.

• Using an estimated population size of 25,459,470, the total percentage of the Australian population confirmed as having been infected currently sits at 0.11%, or one person in every 894.
• Within Victoria, that rises to one in every 325 having contracted the virus at some point

confirmed %>% 
  left_join(recovered, by = c("State", "Country", "date")) %>%
  left_join(deaths, by = c("State", "Country", "date")) %>%
  left_join(active, by = c("State", "Country", "date")) %>%
  bind_rows(vic_aged) %>%
  dplyr::filter(
    date %in% c(dt, dt -1)
  ) %>%
  mutate(
    confirmed = case_when(
      State == "Victoria" ~ confirmed - dplyr::filter(., State == "Victoria (Aged Care)")$confirmed,
      TRUE ~ confirmed
    ),
    recovered = case_when(
      State == "Victoria" ~ recovered - dplyr::filter(., State == "Victoria (Aged Care)")$recovered,
      TRUE ~ recovered
    ),
    deaths = case_when(
      State == "Victoria" ~ deaths - dplyr::filter(., State == "Victoria (Aged Care)")$deaths,
      TRUE ~ deaths
    ),
    active = case_when(
      State == "Victoria" ~ active - dplyr::filter(., State == "Victoria (Aged Care)")$active,
      TRUE ~ active
    ),
    State = str_replace_all(State, "^Victoria$", "Victoria (Non-Aged Care)")
  ) %>%
  group_by(State) %>%
  mutate(
    Increase = c(NA, diff(confirmed)),
    `% Increase` = Increase / min(confirmed),
    recovered = c(NA, max(recovered)),
    deaths = c(NA, max(deaths)),
    active = c(NA, active[2])
  ) %>%
  ungroup() %>%
  pivot_wider(
    id_cols = State, 
    names_from = date, 
    values_from = c(confirmed, recovered, deaths, active, Increase, `% Increase`)
  ) %>%
  dplyr::select_if(function(x){sum(is.na(x)) <= 1}) %>%
  rename_all(
    str_remove_all, pattern = "confirmed_"
  ) %>%
  rename_all(str_remove_all, pattern = "_202[01].+") %>%
  arrange(State) %>%
  bind_rows(
    tibble(
      State = "National Total",
      "{as.character(dt -1)}" := sum(.[[as.character(dt -1)]]),
      "{dt_char}" := sum(.[[dt_char]]),
      Increase = sum(.$Increase),
      recovered = sum(.$recovered),
      deaths = sum(.$deaths),
      active = sum(.$active),
      `% Increase` = Increase / !!sym(as.character(dt - 1))
    )
  ) %>%
  mutate(
    `% Increase` = percent(`% Increase`, accuracy = 0.01),
    `Fatality Rate` = percent(deaths / !!sym(dt_char), accuracy = 0.1),
    `Recovery Rate` = percent(recovered / !!sym(dt_char), accuracy = 0.1),
  ) %>%
  rename(
    Fatalities = deaths,
    Recovered = recovered,
    `Currently Active` = active
  ) %>%
  dplyr::select(
    State, starts_with("20"), ends_with("Increase"), starts_with("Fatal"), starts_with("Recov"), `Currently Active`
  ) %>%
  pander(
    justify = "lrrrrrrrrr",
    caption = paste(
      "*Confirmed cases, fatalities and recoveries reported by each state at time of preparation.",
      "Given that the Victorian outbreak significantly impacted the Federally-run Aged Care facilities,", 
      "Victorian statistics are broken down into those for which the Andrews government is responsible (Non-Aged Care),", 
      "and those __for which the Morrisson government is reponsible (Aged Care)__.", 
      "Detailed statistics on Aged Care are not currently available for other states.*"
    ),
    emphasize.strong.rows = nrow(.)
  )

Confirmed cases, fatalities and recoveries reported by each state at time of preparation. Given that the Victorian outbreak significantly impacted the Federally-run Aged Care facilities, Victorian statistics are broken down into those for which the Andrews government is responsible (Non-Aged Care), and those for which the Morrisson government is reponsible (Aged Care). Detailed statistics on Aged Care are not currently available for other states.

State	2021-01-01	2021-01-02	Increase	% Increase	Fatalities	Fatality Rate	Recovered	Recovery Rate	Currently Active
Australian Capital Territory	118	118	0	0.00%	3	2.5%	114	96.6%	1
New South Wales	4,928	4,947	19	0.39%	54	1.1%	3,197	64.6%	180
Northern Territory	75	82	7	9.33%	0	0.0%	71	86.6%	11
Queensland	1,253	1,255	2	0.16%	6	0.5%	1,224	97.5%	13
South Australia	580	580	0	0.00%	4	0.7%	566	97.6%	10
Tasmania	234	234	0	0.00%	13	5.6%	221	94.4%	0
Victoria (Aged Care)	2,128	2,128	0	0.00%	652	30.6%	1,476	69.4%	0
Victoria (Non-Aged Care)	18,248	18,260	12	0.07%	168	0.9%	18,063	98.9%	29
Western Australia	863	866	3	0.35%	9	1.0%	838	96.8%	19
National Total	28,427	28,470	43	0.15%	909	3.2%	25,770	90.5%	263

Plot of Current Australian Values

ausStatsCap <- "*Current confirmed and recovered cases, along with fatalities for Australia only. Active cases are shown as confirmed cases excluding fatalities and those classed as recovered. Loess curves through all points are shown as continuous lines. Data is only shown from 1^st^ April 2020 as this was the first date of complete data being available. Recovered patient information was also sparse in the early stages of data collection, and as a result estimates of active infections will be a significant underestimate until 6^th^ April. In particular, QLD only began reporting recovered cases on this date. NSW followed a fortnight after this date and as such, only the most recent numbers can be considered as accurate. Below this plot, the same figures can be seen broken down by state.*"

ggplotly(
  confirmed %>%
    dplyr::filter(date <= dt) %>%
    left_join(deaths, by = c("State", "Country", "date")) %>%
    left_join(recovered, by = c("State", "Country", "date")) %>%
    left_join(active, by = c("State", "Country", "date")) %>%
    mutate_at(vars(confirmed, deaths, recovered, active), na_locf) %>%
    group_by(Country, date) %>%
    summarise_at(vars(confirmed, deaths, recovered, active), sum) %>%
    ungroup() %>%
    pivot_longer(
      cols = c(active, confirmed, deaths, recovered),
      names_to = "Status",
      values_to = "Total"
    ) %>%
    arrange(Status, date) %>%
    dplyr::filter(date > ymd("2020-04-01")) %>%
    mutate(
      Status = str_to_title(Status),
      Status = str_replace_all(Status, "Deaths", "Fatal"),
      Status = factor(Status, levels = c("Fatal", "Recovered", "Active"))
    ) %>%
    dplyr::filter(Total > 0) %>%
    rename_all(str_to_title) %>%
    dplyr::filter(Status != "Confirmed") %>%
    ggplot(aes(Date, Total, fill = Status)) +
    geom_col() +
    geom_line(
      data = . %>%
        group_by(Date) %>%
        summarise(
          Total = sum(Total)
        ) %>%
        mutate(Status = "Confirmed"),
      colour = "blue"
    ) +
    scale_fill_manual(
      values = c(
        Active = rgb(0, 0, 0),
        Confirmed = rgb(0, 0.3, 0.7),
        Fatal = rgb(0.8, 0.2, 0.2),
        Recovered = rgb(0.2, 0.7, 0.4)
      )
    ) +
    scale_x_date(expand = expansion(c(0, 0.03))) +
    scale_y_continuous(expand = expansion(c(0, 0.05))) +
    labs("Total Cases")
)

Current confirmed and recovered cases, along with fatalities for Australia only. Active cases are shown as confirmed cases excluding fatalities and those classed as recovered. Loess curves through all points are shown as continuous lines. Data is only shown from 1^st April 2020 as this was the first date of complete data being available. Recovered patient information was also sparse in the early stages of data collection, and as a result estimates of active infections will be a significant underestimate until 6^th April. In particular, QLD only began reporting recovered cases on this date. NSW followed a fortnight after this date and as such, only the most recent numbers can be considered as accurate. Below this plot, the same figures can be seen broken down by state.

ggplotly(
  confirmed %>%
    dplyr::filter(date <= dt) %>%    
    left_join(deaths, by = c("State", "Country", "date")) %>%
    left_join(recovered, by = c("State", "Country", "date")) %>%
    left_join(active, by = c("State", "Country", "date")) %>%
    mutate_at(vars(confirmed, deaths, recovered, active), na_locf) %>%
    dplyr::filter(
      date > ymd("2020-04-01"),
    ) %>%
    arrange(date) %>%
    left_join(ausPops) %>%
    pivot_longer(
      cols = c(confirmed, deaths, recovered, active),
      names_to = "status",
      values_to = "count"
    ) %>%
    dplyr::filter(
      count > 0,
      !(State %in% c("Queensland", "New South Wales") & status == "recovered" & date < ymd("2020-04-06")),
      !(State %in% c("South Australia") & status == "recovered" & date < ymd("2020-04-01")),    
      !(State %in% c("Tasmania") & status == "recovered" & date < ymd("2020-04-02")),
    ) %>%
    dplyr::filter(status != "confirmed") %>%
    mutate(
      rate = 1e6*count/Population,
      rate = round(rate, 2),
      status = str_replace(status, "deaths", "fatal") %>% str_to_title(),
      status = factor(status, levels = c("Fatal", "Recovered", "Active"))
    ) %>%
    rename_all(str_to_title) %>%
    ggplot(aes(Date, Rate, fill = Status, label = Count)) +
    geom_col() +
    geom_line(
      data = . %>%
        group_by(State, Date) %>%
        summarise(
          Rate = sum(Rate),
          Count = sum(Count)
        ) %>%
        mutate(Status = "Confirmed"),
      colour = "blue"
    ) +
    facet_wrap(~State, ncol = 4) + 
    scale_fill_manual(
      values = c(
        Active = rgb(0, 0, 0),
        Confirmed = rgb(0, 0.3, 0.7),
        Fatal = rgb(0.8, 0.2, 0.2),
        Recovered = rgb(0.2, 0.7, 0.4)
      )
    ) +
    scale_x_date(expand = expansion(c(0, 0.03))) +
    labs(y = "Rate (Cases / Million)")
)

Breakdown of individual states. Victorian recovered numbers began to be accurately reported from 22^nd March, with other states gradually providing this information. NSW/QLD recovered cases have only recently begun being reported and up until the most recent dates, recovered/active values were very approximate for these states. The extreme drop for NSW active cases in early June is a function of the changed reporting strategy implemented by NSW Health.

Daily New Cases

ggplotly(
  confirmed %>%
    dplyr::filter(date <= dt) %>%    
    group_by(State) %>%
    mutate(daily = c(0, diff(confirmed))) %>%
    ungroup() %>%
    dplyr::filter(confirmed > 0) %>%
    mutate(
      daily = case_when(
        daily < 0 ~ 0,
        daily >= 0 ~ daily
      )
    ) %>%
    bind_rows(
      group_by(., date) %>%
        summarise(daily = sum(daily)) %>%
        ungroup() %>%
        mutate(State = "All States")
    ) %>%
    group_by(State) %>%
    mutate(
      MA = round(sma(daily, 7), 2),
      MA2 = round(sma(daily, 14), 2),
      `Above Average` = MA > MA2
    ) %>%
    dplyr::filter(date > "2020-03-01") %>%
    ggplot(aes(date, daily)) +
    geom_col(
      aes(fill = `Above Average`, colour = `Above Average`),
      data = . %>% dplyr::filter(!is.na(`Above Average`)),
      width = 1/2
    ) +
    geom_line(aes(y = MA), colour = "blue") +
    geom_line(aes(y = MA2), colour = "black") +
    facet_wrap(~State, scales = "free_y") +
    labs(
      x = "Date",
      y = "Daily New Cases",
      fill = "\nAbove\nAverage"
    ) +
    scale_fill_manual(values = c("white", rgb(1, 0.2, 0.2))) +
    scale_colour_manual(values = c("grey50", rgb(1, 0.2, 0.2))),
  tooltip = c(
    "date", "daily", "MA"
  )
)

Daily new cases for each state shown against the 7-day (blue) and 14-day (black) averages. Days which the 7-day average is above the 14-day average are highlighted in red.

Australian Fatality Rate

inc <- 6
icu <- 11
d <- 7
offset <- icu + d 
minDate <- "2020-04-20"

list(
  confirmed %>%
    dplyr::filter(date <= dt) %>%    
    group_by(date) %>%
    summarise_at("confirmed", sum) %>%
    left_join(
      deaths %>%
        group_by(date) %>%
        summarise_at("deaths", sum)
    ) %>%
    dplyr::filter(
      date > minDate
    ) %>%
    mutate(
      fr = deaths / confirmed,
      type = "No Offset"
    ),
  confirmed %>%
    dplyr::filter(date <= dt) %>%    
    mutate(
      date = date + offset 
    ) %>%
    group_by(date) %>%
    summarise_at("confirmed", sum) %>%
    left_join(
      deaths %>%
        group_by(date) %>%
        summarise_at("deaths", sum) 
    ) %>%
    dplyr::filter(
      date > minDate
    ) %>%
    mutate(
      fr = deaths / confirmed,
      type = glue("Offset ({offset} days)")
    ) 
) %>%
  bind_rows() %>%
  ggplot(
    aes(date, fr, colour = type)
  ) +
  geom_line() +
  scale_x_date(
    expand = expansion(mult = 0, add = 0)
  ) +
  scale_y_continuous(label = percent) +
  labs(
    x = "Date",
    y = "Estimated Fatality Rate",
    colour = "Calculation"
  )

Fatality rate for Australian cases as calculated using two methods. Where no offset is included, the rate shown is simply the number of fatalities divided by the total number of reported cases on the same date. When cases increase during a new outbreak, this will skew the fatality rate lower. An alternative is to use an offset based on the fact the the median time from infection to symptom onset is 6 days, the median time from symptom onset to ICU admission is 11 days, and the median time from ICU admission to mortality is 7 days. When using the offset, the fatality rate is calculated as the number of recorded fatalities on a given date, divided by by the number of cases from 18 days ago. Whilst still flawed this may give a less biased estimate on the true fatality rate, and importantly, will always be higher than the alternative calculation. The intial fatality rate spiked above 30% during the intial outbreak under the offset approach, and as such, data is only shown after 20 Apr, 2020. All times used for estimation the offset were obtained from here

Current Growth Factor

n <- 14
minCases <- 1
cp <- glue(
  "*Growth factor for each State/Territory. 
  __Values are calculated using only locally-acquired cases__.
  In order to try and minimise volatility a {n} day simple moving average was used, in contrast to the 5 day average as advocated [here](https://www.abc.net.au/news/2020-04-10/coronavirus-data-australia-growth-factor-covid-19/12132478).
  This enables assessment of the growth factor over an entire quarantine period.
  This value becomes volatile when daily new cases approach zero as is commonly observed in small populations, and at the end stages of an outbreak. 
  As a result, values are only shown when the {n}-day average of new __locally acquired cases__ exceeds {minCases}.*"
)

gf <- list(
  local %>%
    dplyr::filter(date <= dt) %>%    
    arrange(date) %>%
    group_by(State) %>%
    mutate(
      new = c(0, diff(local)),
      new_ma = sma(new, n)
    ) %>%
    dplyr::filter(local > 0, !is.na(new_ma)) %>%
    mutate(
      R = c(NA, new_ma[-1] / new_ma[-n()]),
      R = case_when(
        is.nan(R) ~ NA_real_,
        new_ma < minCases ~ NA_real_,
        TRUE ~ R
      )
    ) %>%
    ungroup() %>%
    arrange(State),
  local %>%
    dplyr::filter(date <= dt) %>%
    arrange(date) %>%
    group_by(Country, date) %>%
    summarise_at(vars(local), sum) %>%
    ungroup() %>%
    mutate(
      new = c(0, diff(local)),
      new_ma = sma(new, n)
    ) %>%
    dplyr::filter(local > 0, !is.na(new_ma)) %>%
    mutate(
      R = c(NA, new_ma[-1] / new_ma[-n()]),
      R = case_when(
        is.nan(R) ~ NA_real_,
        new_ma < minCases ~ NA_real_,
        TRUE ~ R
      ),
      State = "All States"
    ) %>%
    arrange(State)
) %>%
  bind_rows() %>%
  # dplyr::filter(date > ymd("2020-03-01")) %>%
  ggplot(aes(date, R, colour = State)) +
  geom_ribbon(aes(ymin = 1, ymax = R), alpha = 0.1) +
  geom_hline(yintercept = 1) +
  geom_label(
    aes(label = R),
    data = . %>%
      dplyr::filter(date == max(date), !is.na(R)) %>%
      mutate(R = round(R, 2), date = date + 1),
    fill = rgb(1, 1, 1, 0.3),
    show.legend = FALSE,
    nudge_y = 0.3,
    size = 4
  ) +
  labs(
    x = "Date", y = "Growth Factor"
  ) +
  facet_wrap(~State, scales = "free_x") +
  theme(legend.position = "none") +
  coord_cartesian(ylim = c(0.5, 1.8))#2.1))
gf

Growth factor for each State/Territory. Values are calculated using only locally-acquired cases. In order to try and minimise volatility a 14 day simple moving average was used, in contrast to the 5 day average as advocated here. This enables assessment of the growth factor over an entire quarantine period. This value becomes volatile when daily new cases approach zero as is commonly observed in small populations, and at the end stages of an outbreak. As a result, values are only shown when the 14-day average of new locally acquired cases exceeds 1.

The current 14 day growth factor is 0.97 which gives some degree of confidence that the spread of infections is relatively under control.

Testing Within Each State

tested %>% 
  left_join(confirmed, by = c("State", "Country", "date") ) %>%
  dplyr::filter(date == dt) %>%
  left_join(ausPops,  by = "State") %>%
  bind_rows(
    tibble(
      State = "National Total",
      date = dt,
      Population = sum(.$Population, na.rm = TRUE),
      confirmed = sum(.$confirmed, na.rm = TRUE),
      tests = sum(.$tests, na.rm = TRUE)
    )
  ) %>%
  mutate(
    `Tests / '000` = round(1e3 * tests / Population, 2),
    Positive = confirmed / tests,
    Negative = 1 - Positive,
    isTotal = grepl("Total", State)
  ) %>%
  dplyr::select(
    State, Population,
    Confirmed = confirmed,
    Tests = tests, 
    contains("000"), 
    ends_with("ive"),
    isTotal
  ) %>%
  arrange(isTotal, desc(`Tests / '000`)) %>%
  dplyr::select(-isTotal) %>%
  dplyr::rename(
    `% Positive Tests` = Positive,
    `% Negative Tests` = Negative
  ) %>%
  mutate_at(
    vars(starts_with("%")), percent, accuracy = 0.01
  ) %>%
  pander(
    justify = "lrrrrrr",
    missing = "",
    caption = glue(
      "*COVID-19 testing scaled by state population size.
      Confirmed cases are assumed to be the tests returning a positive result.
      The current numbers available for some states are a lower limit, and as such, the proportion of the population tested is likely to be higher, as is the proportion of tests returning a negative result.*"
    ),
    emphasize.strong.rows = nrow(.)
  )

COVID-19 testing scaled by state population size. Confirmed cases are assumed to be the tests returning a positive result. The current numbers available for some states are a lower limit, and as such, the proportion of the population tested is likely to be higher, as is the proportion of tests returning a negative result.

State	Population	Confirmed	Tests	Tests / ’000	% Positive Tests	% Negative Tests
Victoria	6,629,870	20,388	3,907,338	589.4	0.52%	99.48%
New South Wales	8,117,976	4,947	4,151,443	511.4	0.12%	99.88%
South Australia	1,756,494	580	841,246	478.9	0.07%	99.93%
Northern Territory	245,562	82	82,714	336.8	0.10%	99.90%
Australian Capital Territory	428,060	118	138,054	322.5	0.09%	99.91%
Queensland	5,115,451	1,255	1,485,573	290.4	0.08%	99.92%
Tasmania	535,500	234	144,233	269.3	0.16%	99.84%
Western Australia	2,630,557	866	626,447	238.1	0.14%	99.86%
National Total	25,459,470	28,470	11,377,048	446.9	0.25%	99.75%

R Session Information

R version 4.0.3 (2020-10-10)

Platform: x86_64-pc-linux-gnu (64-bit)

locale: LC_CTYPE=C, LC_NUMERIC=C, LC_TIME=C, LC_COLLATE=C, LC_MONETARY=C, LC_MESSAGES=en_AU.UTF-8, LC_PAPER=en_AU.UTF-8, LC_NAME=C, LC_ADDRESS=C, LC_TELEPHONE=C, LC_MEASUREMENT=en_AU.UTF-8 and LC_IDENTIFICATION=C

attached base packages: stats, graphics, grDevices, utils, datasets, methods and base

other attached packages: QuantTools(v.0.5.7.1), data.table(v.1.13.4), plotly(v.4.9.2.2), pander(v.0.6.3), rvest(v.0.3.6), xml2(v.1.3.2), glue(v.1.4.2), ggrepel(v.0.9.0), scales(v.1.1.1), lubridate(v.1.7.9.2), rlang(v.0.4.9), forcats(v.0.5.0), stringr(v.1.4.0), dplyr(v.1.0.2), purrr(v.0.3.4), readr(v.1.4.0), tidyr(v.1.1.2), tibble(v.3.0.4), ggplot2(v.3.3.2) and tidyverse(v.1.3.0)

loaded via a namespace (and not attached): Rcpp(v.1.0.5), assertthat(v.0.2.1), digest(v.0.6.27), R6(v.2.5.0), cellranger(v.1.1.0), backports(v.1.2.1), reprex(v.0.3.0), evaluate(v.0.14), highr(v.0.8), httr(v.1.4.2), pillar(v.1.4.7), lazyeval(v.0.2.2), curl(v.4.3), readxl(v.1.3.1), rstudioapi(v.0.13), rmarkdown(v.2.6), labeling(v.0.4.2), selectr(v.0.4-2), htmlwidgets(v.1.5.3), munsell(v.0.5.0), broom(v.0.7.3), compiler(v.4.0.3), modelr(v.0.1.8), xfun(v.0.19), pkgconfig(v.2.0.3), htmltools(v.0.5.0), tidyselect(v.1.1.0), fasttime(v.1.0-2), fansi(v.0.4.1), viridisLite(v.0.3.0), crayon(v.1.3.4), dbplyr(v.2.0.0), withr(v.2.3.0), grid(v.4.0.3), jsonlite(v.1.7.2), gtable(v.0.3.0), lifecycle(v.0.2.0), DBI(v.1.1.0), magrittr(v.2.0.1), cli(v.2.2.0), stringi(v.1.5.3), farver(v.2.0.3), fs(v.1.5.0), ellipsis(v.0.3.1), generics(v.0.1.0), vctrs(v.0.3.6), tools(v.4.0.3), Cairo(v.1.5-12.2), hms(v.0.5.3), crosstalk(v.1.1.0.1), yaml(v.2.2.1), colorspace(v.2.0-0), knitr(v.1.30) and haven(v.2.3.1)